A STUDY IN THE VIABILITY OF POLE MOUNTED STRUCTURE FROM MOTION SURVEYS AS GEOMORPHOLOGICAL MONITORING TOOLS

Erosion along stream banks remains a prevalent issue in the Catskills. In recent years, the NYCDEP has played a major role in monitoring and mitigating channel morphological changes at sites acting as major sources of turbidity. This study serves as a proof of concept for the viability of Pole Mounted Structure from Motion (SfM) as a tool for monitoring stream channel morphology. SfM Photogrammetry allows for a 3D reconstruction of an environment using 2D images. Continuous overlap between photos in a set allows for SfM software to accurately project the locations of known tie points between photos.

The NYCDEP already performs yearly drone based surveys of Stream Sediment Turbidity Reduction Projects (STRPs), capturing high definition topographic data at each of these project sites. In our piloting of the methodology, we surveyed 2 sites, the Warner Creek site, outside of Phoenicia, NY, which underwent a restoration project in 2021, and the East Kill landslide site outside of Jewett, NY. Surveys were conducted using consumer grade action cameras mounted at ~5m on poles. Constructed point clouds were compared to publicly available Lidar and DEM data, as well as a preexisting drone based survey of the Warner Creek site, courtesy of SLR Consulting.

Our results support the validity of pole mounted SfM surveys as a monitoring tool. Point clouds we generated showed near perfect alignment to existing point clouds generated through drone based SfM surveys. Our resulting point density of 7mm in between points exceeded data sets used in comparison. This resolution allows for change detection due to surface runoff, as well as larger mass movements. A repeat survey at the East Kill site documented a significant collapse of the landslide face along the creek, resulting in the formation of a 2 meter tall deposit within the channel. Overall, our work demonstrated the viability of pole mounted SfM surveying as a tool for actively monitoring changes in channel morphology.